Evaluation of WIM Auto-Calibration Practices and Parameters

US Transportation Collection2019PDFTech ReportHernandez, SarahBaker, JoshDurandal, Fu ZhangUniversity of Arkansas, Fayetteville. Department of Civil EngineeringArkansas. Department of TransportationArkansas. Department of TransportationUnited StatesAutomatic vehicle identificationCalibrationData collectionGlobal Positioning SystemTrucksWeigh in motionQuality controlWeightTRC1801Final ReportWeigh-in-Motion (WIM) systems capture weight and axle configurations of vehicles using the state highway network and serve as valuable and essential input for evaluating the performance of our transportation infrastructure. To produce accurate weights, sensors must be calibrated at regular intervals. Since on-site calibration can be costly, several states have adopted autocalibration procedures. Auto-calibration is an algorithmic procedure by which weights measured by the WIM sensor are adjusted using reference weights, e.g., assumed front axle weights (FAW) of five-axle tractor-trailers. This project developed a new form of auto-calibration based on Automatic Vehicle Identification (AVI) data, specifically truck Global Positioning System (GPS) data. Two data collection efforts were undertaken to gather WIM, static scale, and video data for algorithm validation resulting in a sample of approximately 500 trucks to use for algorithm performance evaluation. Without any form of auto-calibration, we observed FAW errors ranging from 24% to 85% with Gross Vehicle Weight (GVW) error in the same range. Site-specific tuning of the user specified values required in the Arkansas Department of Transportation (ARDOT) and Minnesota Department of Transportation (MnDOT) auto-calibration algorithms resulted in errors for FAW between 9% and 11% and for GVW between 14% and 18%. A limitation of developing site-specific, user-specific values like FAW references is that it would require very detailed and time consuming data collection efforts. The AVI approach, on the other hand, alleviates the need to perform manual field data collection by leveraging AVI truck tracking technologies such as GPS. The AVI-based method reduces errors to between 10% and 35% for FAW and 16% and 35% for GVW. If auto-calibration using AVI were to replace on-site calibration efforts, ARDOT could realize up to 63%. Future work should examine performance increases resulting from a larger AVI sample, evaluation license plate matching as a basis for AVI data, and the use of AVI data to prioritize WIM site sensor upgrades.125